Viewership and Ratings
Introduction Audience numbers and patterns are critical to the success of a television show. The most critically acclaimed show will soon find itself cancelled if it can't get a sizable enough audience to tune in each week and watch. Viewership patterns are measured by the A.C. Nielsen Company, a Dutch company based in New York. The so-called "Nielsen ratings" are the benchmark by which networks evaluate the success of their shows, and determine whether they should be renewed, cancelled, moved to a different time slot, and more. They are also crucial to attracting advertisers and to setting ad rates, the price an advertiser pays for 10, 15 and 30-second advertising time slots. Every day, Nielsen samples a representative group of families who record, either electronically or in a diary, what they watch. These Nielsen families, who are chosen to represent the demographics of the overall American viewing audience, are used to estimate actual television viewership. From these viewership data, Nielsen is able to calculate viewing patterns, who is viewing and whether they continue viewing once they turn on a show. To accomplish this, Nielsen uses audience sampling, a research method that takes a small sample of people who proportionally represent a larger population along a series of key descriptors, and studies what, and how, they watch television. This research methodology is also commonly used with political polling, such as when CBS News polls sample small numbers of representative voters, and from the results of their poll, predict who might win an election and by how large a margin. Sampling of this kind includes what is known as error, that is, the difference between what the sample tells Nielsen is the number of people viewing a show, and the actual number (known to statisticians as "the truth"). Most statisticians attempt to maintain a margin of error within a 5% band, meaning that for an estimate of 1.0 million viewers, the actual viewership is actually somewhere between .95 (-5%) and 1.05 (+5%) million viewers. The wider the bands, the less accurate the estimate.It should be noted that Nielsen does not report its margin of error, which often indicates that it is quite large, perhaps as much as +/-10%. Nielsen ratings are reported Monday through Friday, the morning after every evening's broadcast (Saturday data are reported with Sunday data on Monday.) By 10:00 am, Pacific time, Nielsen has released "fast national" ratings, numbers based on their electronic data collection; networks will use these for morning-after PR press releases and other time-sensitive activity. By mid-afternoon, Nielsen's "broadcast final" ratings data, which are more complete and more accurate, have been released. It's these broadcast final ratings that the networks pay the most attention. Live + 3 Day and Live + 7 Day ratings follow over longer periods of time (three weeks for the latter.) But what do those figures mean? Measuring the Audience: Viewership, Rating and Share Let's imagine that Nielsen reports one episode of Person of Interest had 10.65 million viewers with a 3.0 rating and a 17 share. Is that good, or should we be worried? It's all in knowing how to read the ratings data. Nielsen uses a variety of approaches to measuring the viewing audience of a given show. The three most common methods are: * Viewership: An estimate of the actual number of people watching a show. This is the simplest measure -- how many people does Nielsen estimate are actually tuned in to Person of Interest, and do they stay tuned in? To measure this, Nielsen reports average viewership for an entire episode, as well as in half-hour increments. That allows Nielsen to determine how many viewers tune in and how many viewers who tune into the first half of the episode are still watching during the second half of the episode, which is particularly important during a show's first few episodes. This methodology also allows Nielsen to examine viewership habits from hour-to-hour, and estimate audience retention. In the case of Person of Interest, CBS is interested in how much of the NCIS:LA (9:00 pm) audience POI retains at 10:00pm. If NCIS:LA had 13.27 million viewers the same night POI had 10.65 million viewers, POI retained 80.3% of its lead-in audience. One of the biggest problems with understanding viewership figures is that the potential number of people in the audience for a program varies from night-to-night and from week-to-week. One week, 10.65 million viewers might be 15% of the number of people actually watching television; another week, it might be 24%. What's more, a special event, such as the World Series may pull off a large part of the audience for one or two nights. Day of the week and time of day are important, too: the potential TV audience is larger on a Thursday than on a Tuesday, and at 9:00 than at 10:00, which also influence how we interpret, and especially compare viewership figures from week-to-week, from season to season, or between shows. That makes viewership figures alone a poor measure of how successful a show actually is. To combat this problem, Nielsen uses two additional statistics which mathematically equate viewership numbers so we can make some comparisons: rating and share. * Rating: In the United States, there are currently 115.6 million households with one or more television sets, what Nielsen calls "television households." A rating point represents 1% of the television households viewing a show, or 1.156 million viewers. Thus, a 3.0 rating would be 3 x 1.156 = 3.468 million households. Use of ratings allows Nielsen to measure not only the total audience but audience subgroups by age, gender or ethnicity. The most prized group, and the rating we generally see reported, is the 18-49 age group, the group Nielsen and most advertisers believe are the most active spenders (a supposition that is increasingly being called into question.) Consequently, our 3.0 rating above would indicate 3.47 million households with at least one 18-49 year old, or 3% of the potential households with an 18-49 year old, tuned in to POI. * Share: The audience share is the percentage of televisions in use (that it, turned on) that tuned into a television show. A 17 share would indicate that 17% of televisions in use are watching Person of Interest. Although the least reported statistic, it is the most telling because it gives us the percentage of the people actually watching TV (versus potential audience) who were watching POI. Rating Time Periods Nielsen reports viewership, rating and share for multiple time periods. Originally, they reflected live viewership only. With the advent of the VCR, and now the DVR and other electronic media, Nielsen has begun to collect data on time shifted viewership, and reports data for three time blocks: Live + Same Day audience (viewers who watch live or within 24 hours), Live + 3 Day and Live + 7 Day (time shifted or "DVR" viewership over the following three, then seven days.) Nielsen began to collect time-shifted data in 2005, but advertisers were reluctant to use these data until around 2010, so it's only recently we've begun to see these data routinely reported. Nielsen also collects viewership data for both first-run broadcasts and reruns. For shows that run twice in an evening, such as seen with many cable summer shows, ratings data for each broadcast are gathered and reported separately. Broadcast Sweeps Three times a broadcast season, in November, February and May, viewership and ratings become particularly important. The ratings during these periods of time known as "sweeps", are the basis for price networks will charge for advertising time: the higher the ratings, the more a network can charge for 10, 15 and 30 second advertising time slots and the more prestigious advertisers they will attract. Reruns, which result in far lower viewer numbers, are rare during these periods. Sweeps are all about new, high profile programming. These high-stakes blocks of time are when shows roll out episodes designed to draw the largest possible number of viewers. Advertising campaigns that accompany these episodes are also produced to pull viewers. Although Live + Same Day, Live + 3 Day and Live + 7 Day viewership numbers are all important, live numbers are critical, and the networks will do what they can to get you in front of your TV when a sweeps episode is broadcast (when you will also watch the advertisements) rather than watching it on your DVR a couple days later (when you can skip over the ads.) Sweeps episodes will often feature resolution of long-running story lines, multi-episode story arcs, major events such as weddings, the return of a departed cast member, or "stunt" casting of high-visibility actors, all to induce viewers to watch live. Most series' season finale will fall late in the May sweeps period. The recent "Endgame" arc, which concluded Carter's search for the leaders of HR, is a classic example of how a sweeps period story arc will work. In the episodes prior to , the events leading to the arc itself played out, concluding with Terney's identification of Alonzo Quinn as the head of HR in , broadcast the first week of sweeps. With the groundwork in place, CBS rolled out an intense, and intensive, advertising campaign centering on the arc, giving it a title, leading us to believe Fusco would die, and (hopefully) grabbing our attention. Over the next three weeks, we watched breathlessly as the story played out, then spent hours afterward talking about it on our forum -- exactly what Bad Robot and CBS wanted. And hopefully, we also bought the products that were advertised during those expensive ad breaks. For CBS, that's their endgame: advertising dollars equal profits. The final episode of the sweeps period generally concludes with an open ending that leaves us asking, "what happens next?" and keeps us coming back for more. Remember and , that stunned us, then kept us on the edge of our seats as the new Samaritan story line began. We hung in over the holiday break waiting to find out whether John would return, what Decima was up to, and what Control would do to Finch and his college friend. That's what sweeps are all about: they keep us watching and keep us watching those ads. Are the ads really that important? Ask yourself: how long after the Super Bowl were you talking about the new ads, even if you didn't watch the game? And more importantly, what products did you buy as a result? 'Sweeps Periods for the 2014-2015 Broadcast Season' Sweeps run quarterly over four weeks, beginning on a Thursday and ending on a Wednesday. * October 2014: October 30 – November 26, 2014 * February 2015: January 29 – February 25, 2015 * May 2015: April 23 – May 20, 2015 Note: A fourth sweeps period, during July, focuses on daytime programming, new network programming and cable programming. POI's ratings during this period are reported, but do not determine the future of the program because it is in reruns. However, ratings do determine summer ad rates and whether POI keeps its time slot during the summer or gives it up to a summer program. Scheduling and Time Slot The American television "prime time" schedule runs from 8:00 pm - 11:00 pm in the Pacific (Los Angeles), Alaskan (Anchorage) and Eastern (New York) time zones; with programming running one hour earlier in the Mountain (Denver), Central (Chicago) and Hawaiian (Honolulu) time zones. As a rule of thumb, situation comedies and lighter dramas run from 8:00 - 9:00 pm, which is considered the family hour. Heavier and more adult content is placed at 9:00, with the most adult, violent or dramatic content at 10:00 pm. Individual networks place programs based not only on content, but on their ability to draw and audience, follow another program or lead into a program. Person of Interest is currently broadcast in the U.S. on Tuesday at 10:00 pm Pacific time, after two seasons on Thursday at 9:00 pm. As of Season 3, Person of Interest is also broadcast concurrently with the U.S. on CTV in Canada; additionally many cities along the U.S./Canadian border can receive both U.S. and Canadian broadcasts of the show. In its second season, POI served as a strong lead-in for new drama Elementary, before moving to Tuesday, where it was expected to successfully occupy a long problematic time slot following the highly successful programs NCIS and NCIS: Los Angeles. This move also allowed the producers latitude to explore darker topics, and was more appropriate to the level of violence displayed by the show. In September 2015, it will follow new NCIS spin-off NCIS: New Orleans, indicating CBS has confidence in its ability to draw an audience without the support of a strong lead-in program. Season 1 Season 1 Viewership'http://www.seriesmonitor.com/personofinterest/season1.html Season 1 was broadcast on Thursday at 9:00 E/P time (one hour earlier in the Central and Mountain time zones). POI followed a one-hour block of sitcoms: the established hit ''The Big Bang Theory at 8:00 followed by the quickly cancelled How to Be a Gentleman, and later Rules of Engagement. The first season received strong critical acclaim, including some of the highest test audience ratings for a CBS pilot on record. 9:00 pm is typically the broadcast hour with the largest potential audience, and Thursday, along with Sunday, traditionally have the largest audience by day. The series remained in this time slot through its second season. CBS made several experimental adjustments to the 8:00 time block to improve POI's lead-in audience, including a try-out of the sitcom Rob and airings of Big Bang Theory at 8:30. Regardless of its lead-in, POI consistently grew in viewership, as a 9:00 drama would be expected to do. {| width="100%" style="border: 3px solid #999; padding: 1px; color:#000;" ''Person of Interest'' -- Season 1 Viewers Live + Same Day (L+SD) and Live + 7 Day (L+7D) in millions ['''S = sweeps episode] Episode L+SD L+7D Episode L+SD L+7D Viewership for POI grew steadily over its first season, ranging from 11.57 ( ) to 15.67 ( ) million viewers, with an average live audience of 13.30 million viewers, and an average Live + 7 Day audience of 16.28 million, growing 18% from its live broadcast. Critically, Person of Interest retained 94% of its premiere audience for week two, and 87% for week three, quickly indicating CBS had a hit on its hands. Median growth from live to Live + 7 Day was 18%; the greatest growth was for at 21%. The largest audience including time-shifted data was 18.88 million (growing 17% from live) for . {| width="80%" Season 2 Season 2 Viewership'''http://www.seriesmonitor.com/personofinterest/season2.html {| width="100%" style="border: 3px solid #999; padding: 1px; color:#000;" ''Person of Interest'' -- Season 2 Viewers Live + Same Day (L+SD) and Live + 7 Day (L+7D) in millions ['''S = sweeps episode] Episode L+SD L+7D Episode L+SD L+7D {| width="80%" Season 3 Season 3 Viewership'http://www.seriesmonitor.com/personofinterest/season3.html In Season 3, CBS moved the show to Tuesday at 10:00 pm E/P, a time slot where it has traditionally had difficulty placing a successful show to follow its major hits ''NCIS and NCIS:LA. Tuesday often has a smaller potential audience, and typically, viewership drops between 9:00 and 10:00 pm, when the audience turns to local news or begins to head to bed. However, growth from Live + Same Day to Live + 7 day (so-called DVR) viewership tends to increase significantly. Audience retention from 9:00 to 10:00 becomes important, and the most successful shows retain the bulk of their lead-in show's audience. The CBS broadcast schedule for Tuesday begins with their ratings bonanza NCIS, followed by the far less popular NCIS: LA. POI has been the first show to successfully occupy the 10:00 pm slot since The Good Wife in 2010-11. Four series subsequently failed in the slot before POI moved in. The February sweeps period was abbreviated this season because of the 2014 Olympic Winter games, broadcast on NBC. Although the games fell across two sweeps weeks (February 7 - 23), networks rarely broadcast original programming against NBC's prime time coverage. CBS reran and during the second and third weeks of sweeps, separating its two original sweeps broadcasts, and by three weeks. {| width="100%" style="border: 3px solid #999; padding: 1px; color:#000;" ''Person of Interest'' -- Season 3 Viewers Live + Same Day (L+SD) and Live + 7 Day (L+7D) in millions ['''S = sweeps episode] Episode L+SD L+7D Episode L+SD L+7D {| width="80%" Season 4 Season 4 Viewership Note: There is a 2-3 week lag on L*7 data availability. {| width="100%" style="border: 3px solid #999; padding: 1px; color:#000;" Person of Interest -- Season 4 Viewers Live + Same Day (L+SD) and Live + 7 Day (L+7D) in millions [S''' = sweeps episode] '''Episode L+SD L+7D Episode L+SD L+7D Graphs {| width="80%" style="border: 1px solid #999; padding: 1px; color:#000;" Person of Interest Season 1 {| width="80%" style="border: 1px solid #999; padding: 1px; color:#000;" Person of Interest Season 2 {| width="80%" style="border: 1px solid #999; padding: 1px; color:#000;" Person of Interest Season 3 *There is a 2-3 week lag on L*7 data availability. {| width="80%" style="border: 1px solid #999; padding: 1px; color:#000;" Person of Interest Audience by Season Comparisons {| width="80%" style="border: 1px solid #999; padding: 1px; color:#000;" Person of Interest Ratings by Season Comparisons Other viewer data will be added at a later time. References